Behavioral, Cognitive, and Affective Predictors

This study examined multidimensional predictors of child conduct problems in the context of parent-child interactions. Seventy-one children aged eight to twelve participated with their mothers. Mother-child dyads engaged in an experimental clean-up task designed to elicit conduct problems. Hierarchical regression analyses determined the predictive utility of behavioral, cognitive, and affective variables. A regression model that included parent predictors (maternal coercion, perceived power, and anger) and child predictors (perceptions of maternal support, negative attributions for mothers’ behavior, and anger) accounted for 50 percent of the variance in child noncompliance during the clean-up task. Maternal coercion (negative talk) was the most significant predictor; however, incremental variance accounted for by child variables (particularly child anger) provides evidence for mother-child effects. These data highlight the importance of examining interventions that target relational parent-child variables rather than maternal behaviors alone. Further research is needed to clarify the direction of transmission (unidirectional versus bidirectional) of maladaptive parent-child processes and well as to evaluate the extent to which interactive therapies can ameliorate older children’s conduct problems.